a highly reliable, modular, redundant and self-monitoring psu...

Acta Polytechnica Hungarica Vol. 17, No. 7, 2020

– 233 –

A Highly Reliable, Modular, Redundant and

Self-Monitoring PSU Architecture

Bertalan Beszédes, Károly Széll, György Györök

Alba Regia Technical Faculty, Óbuda University

Budai út 45, H-8000 Székesfehérvár, Hungary

{beszedes.bertalan; szell.karoly; gyorok.gyorgy}@amk.uni-obuda.hu

Abstract: The production of highly reliable, electronic devices is a source for significant

environmental emissions and energy consumption. A modern, cost-effective and modular

design can enhance product maintainability and lifetime. Many end-users would certainly

be willing to devote more resources (money) for a device they use, if, in return, they could

extend the life of the device. This paper introduces the architecture for a high-reliability,

modular, end-user-configurable, redundant power supply, based on these principles.

Keywords: redundant; robust; self-monitoring; embedded system; modular PSU; high

reliable

1 Introduction

Despite the arrival of many renewable energy sources, the vast majority of the

world's energy supply is still provided by fossil fuels, the extraction and use

involves the release of large amounts of greenhouse gases into the atmosphere.

Therefore, improving energy efficiency is currently also the most effective means

of trying to fight climate change. While the world's energy efficiency is improving

[1], so is the amount of wasted energy, decreasing. As a result of the growth of

Earth's population and global economy, humanity's total energy consumption

(global primary energy demand) continues to rise. And because of the

predominance of fossil fuels, polluting energy sources in the global energy mix, it

is also leading to an increase in greenhouse gas emissions, which have recently

grown at the fastest pace since 2013.

Technological advances have made it possible to increase energy efficiency,

which significantly reduces emissions while increasing energy use. Energy

efficiency could be improved at a much higher rate, with technologies that are

already available, but rarely used [2] [3].

B. Beszédes et al. A Highly Reliable, Modular, Redundant and Self-Monitoring PSU Architecture

– 234 –

According to the International Energy Agency (IEA), the potential is enormous,

and by taking advantage of it alone, we could stop the rise of greenhouse gas

emissions after 2020. However, according to the latest surveys, the world is

moving further and further away from this goal [4] [5].

According to the International Energy Agency's Efficient World Strategy (EWS),

at least a 3% improvement would be needed, each year, to meet global climate and

sustainability goals. The 3% has only been realized once, in 2015 and the pace has

slowed gradually, thereafter.

By using cost-effective technologies [6], the pace of energy efficiency

improvement can be increased to a much higher level. Design principles like

modular device design, maintainability and traceability can serve this goal.

Digitalization and remote supervision systems are closely linked to these features

[7].

The continuous development of end-user technological capabilities also supports

the need for modular, easy-to-maintain designs [8]. There is a niche market for

manufacturers to ensure user-friendliness and repair-ability of civil and industrial

devices - in exchange for some extra cost. The authors hope that the modular

design and the installation and repair manuals of the devices will come back into

vogue.

In terms of robustness of such systems successful results have been achieved in

the field of fault diagnosis and fault tolerant techniques [9] - [11]. Basically, we

can define the following categories for fault diagnosis methods:

Model-based [12]–[14]

The outputs of the system-model and the outputs of the real system is

compared

Signal-based [15]–[18]

A diagnostic decision is made based on the measured signal

Knowledge-based [19]–[21]

A large volume of historic data is needed

Hybrid and active [22]–[24]

Combination of the previous methods based on their advantages

The basic fault types are as follows [25] [26]:

Actuator fault

Sensor fault

Plant fault

In the literature, several practical solutions can be found [27] [28].


– 235 –

2 Architecture

2.1. Power Supply Unit

2.1.1. Modular Power Supply Unit

Figure 1 shows a block diagram of a battery-powered modular power supply

(PSU) controlled by a microcontroller (MCU). The external source of energy can

be the electricity grid or renewable energy source. The battery charger is

responsible for properly charging and discharging the energy storage unit. The

function of a DC/DC converter is to ensure the voltage level of the battery or

external power source to meet customer needs.

In case of AC external power source or a load that requires AC power supply, the

proper battery charger and AC/DC, DC/AC or AC/AC converter has to be applied.

External power source

Battery charger

DC / DC Load

Rechargable battery

Figure 1

Modular power supply structure

2.1.2. Redundant Power Supply Unit

For single-redundant PSU (see Figure 2), if one of the PSU fails, the backup PSU

takes over the task. This means that only one PSU is working at a time and that it

supplies 100% of the required electricity for the powered system. In this case, the

backup power supply is out of service or under test. This design ensures that the

power supply is fault tolerant. This mode is also called, hot-stand-by mode.


– 236 –

Capacitor

PSU 1 PSU 2

Switching Matrix

PSUmicrocontroller

control

measure

Load


measure

Figure 2

Redundant power supply structure with supervisor MCU

Another solution is the load-sharing mode, where power supplies share the load

power. If there are more than two power supplies in the system and one is out of

service for failure, replacement, or testing, the remaining power supplies will

share the total load current equally.

For example, if there are four redundant PSUs in the power supply system and one

of them, for the above mentioned reasons, goes out of service, the power supplies

that are still in operation, will distribute the load, so, for example, the single units

would provide 33% instead of 25% of the load current. All power supplies would

be able to provide full load current if left alone in the power supply system.

2.1.3. Redundant Modular PSU

On the Figure 3, it can be seen, the capacitor supplies power to the

microcontroller, which supervises the PSU, the connected load, and other optional

modules, when the power supply is temporarily interrupted due to replacement of

the redundant modules.


– 237 –

Battery charger 1

Rechargable Battery 1

Switching & Measuring

Matrix

Battery charger 2

Switching, Measuring

Matrix


Matrix

Rechargable Battery 2

Load

CapacitorTest Load


PSU MCU

DC / DC 1 DC / DC 2


Matrix

Figure 3

Simplified redundant modular power supply unit structure


– 238 –

2.1.4. Microcontroller

Power supplies usually include a microcontroller that affects the power supply and

provides measurement, logging, communication, etc. functions as well.

In the case of modular power supplies, the power supply control microcontroller

and the tightly-coupled components may be provided as separate modules or may

be part of the motherboard.

For redundant power supplies or redundant modular power supplies, the

microcontroller for power supply control extends to monitoring, testing,

evaluating the power supplies or power supply modules, and controlling switching

matrices.

2.1.5. Switching Matrix

The microcontroller also controls the switching matrices that connect the power

modules. The function of the switching matrices is to provide energy flow

between the power modules in a reconfigurable manner. The switching matrices

can be used to connect and disconnect redundant power modules, including

replacing redundant power modules.

When replacing power modules, switching multiple redundant power modules in a

way that would lead to a short circuit should be avoided. The first step is to

disconnect the currently active module, after which the redundant module can be

turned on. The process results in a short-term power line break, but with the

addition of energy storage devices (buffer capacitors), the power supply is

continuously maintained, and transient-low switching is possible. In the

experimental setup galvanically isolated relay modules were used. In case of the

end product it is recommended to use modern technology based semiconductor

switching elements [29]–[31].

2.2. Measuring Method

In redundant fault tolerant systems, some basic features are needed for proper

operation. In hot-stand-by mode or load-sharing mode, the embedded monitoring

system must be able to notify a supervisor and a control system of the actual status

or failure of power supplies and modules.

The embedded monitoring system must be able to monitor power supplies and

modules. Depending on the various aspects of error detection, this may occur

using normal operational load or dummy load. Both the normal operation and the

test operation must be measured with active and standby power modules. This can

be done by swapping the power modules or by intermittently disabling them, as in

a test procedure.


– 239 –

Redundant fault tolerant power supply systems must be provided with continuous

power supply even when defective modules are replaced. The hot-plug-in function

ensures smooth operation of the powered system.

2.2.1. Module Measurement

During the measurement of the modular power supply, the module's efficiency,

temperature, input and output voltage and current values and waveforms must be

monitored. The measured values should be transmitted to the microcontroller for

further processing and storage.

Measured module

Current measurement

Current measurement

IN OUT

Iin

Uin

UDS

Uout

Iout

Voltage measurement

Voltage measurement

GND GND

Figure 4

Module measurement scheme

Voltage measurement, if higher than the reference voltage of the analog digital

converter of the microcontroller, is carried out by a voltage divider made of high

precision and stable elements. There is even the possibility of using an

optocoupler, but the brightness degradation of its built-in semiconductor LED,

typically in the infrared range, can over time falsify the measurement.

Current measurement can be accomplished by measuring or calculating the

differential voltage across Shunt resistors with lower cost. In order to reduce the

number of test cables or to measure higher current values, it is also possible to use

hall sensors, which have the disadvantage of higher costs.

For lifetime prediction, the voltage difference of the semiconductor drain-source

shown in Figure 4 is also measured when the MOSFET is open. When the

MOSFET is closed, the drain-source voltage can easily be higher than the input of

the operational amplifier used to measure the voltage difference, so the

measurement must be constructed with a galvanically isolated operational

amplifier.


– 240 –

2.2.2. Analog Measurements

The measurement and evaluation of the power modules can be accomplished by

using integrated hardware elements, which support the measurement. They can

also detect overcurrent, overvoltage, voltage drop, voltage fluctuation, instability,

voltage loss and other anomalies.

The configuration provides additional options for controlling the output voltage

and for detecting common differences listed below:

After a positive voltage spike, the output voltage returns to normal

After a positive voltage spike, the output voltage will remain above

normal (see Figure 5)

The output voltage remains stably higher than normal, after a voltage

surge

After a negative voltage spike, the output voltage returns to normal (see

Figure 6)

After a negative voltage spike, the output voltage remains below normal

The output voltage gradually decreases

The output voltage fluctuates below normal level

The output voltage remains stable below a normal level after a voltage

drop (see Figure 7)

The output voltage becomes unstable (see Figure 8)

Figure 5

A positive voltage spike, and the output voltage

will remain above normal level

Figure 6

Negative voltage spike the output voltage will

remain normal


– 241 –

Figure 7

The output voltage remains stable below a normal

level after a voltage drop

Figure 8

Unstable output voltage

2.2.3. Multiplexing Analog Lines

The microcontroller that monitors the power supply and controls the switching

matrices has a limited number of terminals and is required to expand due to the

large number of measurement and control signals. Extending the number of

control outputs is easily accomplished with a serial I/O extender IC. The analog

signals to be measured are coupled via an analog multiplexer to the ADC

terminals of the microcontroller.

Choosing a more advanced microcontroller eliminates the need for external

hardware, with sufficient software (multiplexing the input terminals to the ADC

peripheral) to implement the solution presented. If the microcontroller has

multiple internal ADC peripherals, it is recommended to measure the same analog

signals with the same ADC modules - to avoid measurement errors due to

differences in measurement peripherals. If the microcontroller has one internal

ADC periphery, a sample and hold (SH) circuit is necessary to be used.


– 242 –

Analog

MUX

Select

channel

Out

...

MicrocontrollerADC

I2C

Uin1

Iin1

Uout2

Iout2

Uin2

I/O Extender

Iin2

Uout1

Iout1

Switching matrix

control

SH

SH

SH

SH

SH

SH

SH

SH

Figure 9

Hardware measurement and control scheme

2.2.4. External Interrupt Subsystem

Most of the time, the microcontroller controlling the power supply is in sleep

mode. During sleep mode or when executing an instruction, you may not be able

to detect unexpected voltage fluctuations in the power supply. Due to the

architecture of the interrupt request system, which is designed as an external

hybrid circuit, it is able to continuously monitor the output voltage state and, for

example, to request an interrupt from the microcontroller in case of voltage

fluctuation outside the specified limit.

Unlike the block diagram shown in Figure 1, the use of a built-in comparator as

the internal periphery of the microcontroller controlling the power supply is

recommended, in which case the reference voltage can be set as a register content

by software. The internal comparator peripheral of the microcontroller also has the

ability to request an interrupt (see Figure 10).


– 243 –

Uin. Uout, Iin, Iout, UDS

Uin min.

prog. reference

Uin max.

prog. reference

Analog MUX

VDD

I2C BUS

Select

channel

Out

...

...

...

...

Microcontroller

INT

ADCI2C

Figure 10

Block diagram of the external interrupt system

3 Realization

3.1. Main Program

The monitoring system calculates the primary and secondary power of the

modules by measuring the input and output voltages and currents of the power

supply modules. Based on these measurements, it calculates the efficiency of the

power supply modules. If this value is below a predetermined level, or based on

several measurements, it can be determined from the stored results that the

condition of the unit is deteriorating (the efficiency value drops below a certain

level), the monitoring system will send an error message to the monitoring system

and jumps to a subroutine.

The monitoring system must process a significant amount of data. The cost-

effective microcontroller-observed monitoring system can measure only one point

at a time, and the analog-to-digital conversion takes a finitely long time. Power


– 244 –

modules have relatively large number of measurement points. The frequency of

each measurement and the accuracy of analog-to-digital conversion (measurement

time) can be dynamically changed for efficient operation.

In case the error of a measurement point shows only a slight deviation from the

ideal, it can be checked at a lower frequency and with less accuracy, so

monitoring the measurement point takes only a small amount of time in the cycle.

If the error of the measuring point increases, for example, on the basis of a look-

up table, it is recommended to increase the frequency and the accuracy of the

measurement. If the rate of change of the error at the measurement point increases,

it is recommended to further increase the measurement frequency and accuracy of

the measurement point to monitor the change. Measured and stored data can be

used to perform fault prediction functions.

Figure 11

The model circuit

3.2. Operation Modes

The software supports user settings. The user has the ability to select the mode of

operation and to weigh the following considerations (see Figures 11 and 12).


– 245 –

In case the user wants to maximize the life of the device, because maintenance is

difficult, access to the device is difficult or the goal is to reduce the carbon

footprint, you choose Swapping mode with reduced maximum charging and

discharging currents. The swapping modules subroutine switches between

redundant elements primarily based on the efficiency of the modules, but it can

also change based on the temperature of the modules (the temperature of the

active module increases), thus saving parts from heat stress and faster aging.

With advanced user settings, it is possible to fine-tune the above-mentioned

parameters, customize reference levels, hysteresis values and timings.

If higher reliability is the main goal, it is recommended to activate Simultaneously

running mode. In this case, the redundant modules run at 50-50% load. The heat

load is higher than in the previous case. If one module fails, the other takes over

100% of the load with a smaller transient.

This type of control is recommended for powering easily maintainable equipment.

In the case of a module failure, the failure of one of the modules increases the

likelihood of a failure of the module remaining in the system, and in the case of

the failure of the remaining 100% load module there are no spare modules in the

system. The 100% load should only be tolerated by the module remaining in the

system until maintenance, so we may use a lower power margin than in Swapping

or Backup mode.

The third mode of operation is a classic Backup Mode. The module that builds the

primary power supply will operate until it fails, after which the backup module

will take over the load. In this case, the backup module may remain reliable for a

long time due to higher performance margin compared to the Simultaneously

running mode, although it does not include a spare element after the failure of the

backup module.

This control mode is used to drive the primary active module till failure or till a

predetermined degradation of efficiency, thus the long-term power consumption

of this control mode will be the highest.

In all three cases, especially in the Backup Mode, it is important to periodically

test the modules and determine their functionality. Measurements can be made

with the programmatically variable load or with relatively short operation of the

inactive module, until reaching the operating temperature, to perform

measurements during normal operation.


– 246 –

Swapping modeSimultaneously

running mode

Difference

higher than

allowed

Swap modules

Main module is

out of order

Swap to backup

module

Measured

parameters lower

than the error

limit

One of the

modules is out of

order

Turn off the

faulty module

Send error

message

Backup mode

Y

N

Y Y

Y Y Y

N N

N N N

Exit switching subroutine

Start switching subroutine

Figure 12

Mode selector algorithm

Conclusions

The presented redundant power supply, greatly increases the reliability of the

device. A cost-effective solution that monitors itself and a modular architecture

that greatly enhances upgrades and maintainability, which can significantly

increase the life of the equipment, thus, reducing global emissions/waste. The

Authors believe that the presented system architecture, can be successfully

implemented for both Civil and Industrial applications and especially for

applications that demand a high level of reliability.


– 247 –

Acknowledgement

The Authors wish to thank the support to the Arconic Foundation. Bertalan

Beszédes thankfully acknowledge the financial support by the ÚNKP-19-3 New

National Excellence Program of the Ministry for Innovation and Technology,

GINOP-2.2.1-15-2018-00015 and GINOP-2.2.1-15-2017-00098 projects. Károly

Széll thankfully acknowledge the financial support of this work by the Hungarian

State and the European Union under the EFOP-3.6.1-16-2016-00010 and GINOP-

2.2.1-15-2017-00073 projects.

References

[1] M. Kolcun, A. Gawlak, M. Kornatka, and Z. Čonka, “Active and Reactive

Power Losses in Distribution Transformers,” Acta Polytech. Hung., Vol.

17, No. 1, pp. 161-174, 2020, doi: 10.12700/APH.17.1.2020.1.9

[2] “Energy Efficiency 2019,” IEA, Paris, 2019. Accessed: Mar. 26, 2020

[Online] Available: https://webstore.iea.org/download/direct/2891

[3] “Energy Efficiency Indicators 2019,” IEA, Paris, 2019. Accessed: Mar. 26,

2020 [Online] Available:

https://webstore.iea.org/download/direct/2707?fileName=Energy_Efficienc

y_Indicators_2019_Highlights.pdf

[4] World Energy Balances 2019. Paris: IEA, 2019

[5] World Energy Statistics 2019. Paris: IEA, 2019

[6] G. Györök, Programozható analóg áramkörök mikrovezérlő környezetben,

Vol. 1, Székesfehérvár: Óbudai Egyetem, 2013

[7] É. Hajnal, “Big Data Overview and Connected Research at Óbuda

University Alba Regia Technical Faculty,” in AIS 2018 - 13th International

Symposium on Applied Informatics and Related Area, 2018, pp. 1-4

[8] T.-M. I. Băjenescu and M. I. Bazu, Reliability of Electronic Components: A

Practical Guide to Electronic Systems Manufacturing. Berlin Heidelberg:

Springer-Verlag, 1999

[9] Z. Gao, C. Cecati, and S. X. Ding, “A Survey of Fault Diagnosis and Fault-

Tolerant Techniques—Part I: Fault Diagnosis With Model-Based and

Signal-Based Approaches,” IEEE Trans. Ind. Electron., Vol. 62, No. 6, pp.

3757-3767, Jun. 2015, doi: 10.1109/TIE.2015.2417501

[10] Z. Gao, C. Cecati, and S. X. Ding, “A Survey of Fault Diagnosis and Fault-

Tolerant Techniques—Part II: Fault Diagnosis With Knowledge-Based and

Hybrid/Active Approaches,” IEEE Trans. Ind. Electron., Vol. 62, No. 6,

pp. 3768-3774, Jun. 2015, doi: 10.1109/TIE.2015.2419013

[11] W. Kong, Y. Luo, Z. Qin, Y. Qi, and X. Lian, “Comprehensive Fault

Diagnosis and Fault-Tolerant Protection of In-Vehicle Intelligent Electric


– 248 –

Power Supply Network,” IEEE Trans. Veh. Technol., Vol. 68, No. 11, pp.

10453-10464, Nov. 2019, doi: 10.1109/TVT.2019.2921784

[12] J. Hu, J. Wang, J. Zeng, and X. Zhong, “Model-Based Temperature Sensor

Fault Detection and Fault-Tolerant Control of Urea-Selective Catalyst

Reduction Control Systems,” Energies, Vol. 11, No. 7, p. 1800, Jul. 2018,

doi: 10.3390/en11071800

[13] P. Szcześniak, G. Tadra, and Z. Fedyczak, “Model Predictive Control of

Hybrid Transformer with Matrix Converter,” Acta Polytech. Hung., Vol.

17, No. 1, pp. 25-40, 2020, doi: 10.12700/APH.17.1.2020.1.2

[14] L. R. Neukirchner, A. Magyar, A. Fodor, N. D. Kutasi, and A. Kelemen,

“Constrained Predictive Control of Three-PhaseBuck Rectifiers,” Acta

Polytech. Hung., Vol. 17, No. 1, pp. 41-60, 2020, doi:

10.12700/APH.17.1.2020.1.3

[15] H. Chen and S. Lu, “Fault Diagnosis Digital Method for Power Transistors

in Power Converters of Switched Reluctance Motors,” IEEE Trans. Ind.

Electron., Vol. 60, No. 2, pp. 749-763, Feb. 2013, doi:

10.1109/TIE.2012.2207661

[16] N. M. A. Freire, J. O. Estima, and A. J. Marques Cardoso, “Open-Circuit

Fault Diagnosis in PMSG Drives for Wind Turbine Applications,” IEEE

Trans. Ind. Electron., Vol. 60, No. 9, pp. 3957-3967, Sep. 2013, doi:

10.1109/TIE.2012.2207655

[17] M. Shahbazi, E. Jamshidpour, P. Poure, S. Saadate, and M. R. Zolghadri,

“Open- and Short-Circuit Switch Fault Diagnosis for Nonisolated DC–DC

Converters Using Field Programmable Gate Array,” IEEE Trans. Ind.

Electron., Vol. 60, No. 9, pp. 4136-4146, Sep. 2013, doi:

10.1109/TIE.2012.2224078

[18] A. M. Stanisavljević, V. A. Katić, B. P. Dumnić, and B. P. Popadić, “A

Comprehensive Overview of Digital Signal Processing Methods for

Voltage Disturbance Detection and Analysis in Modern Distribution Grids

with Distributed Generation,” Acta Polytech. Hung., Vol. 16, No. 5, Aug.

2019, doi: 10.12700/APH.16.5.2019.5.8

[19] X. Xiang, C. Yu, and Q. Zhang, “On intelligent risk analysis and critical

decision of underwater robotic vehicle,” Ocean Eng., Vol. 140, pp. 453-

465, Aug. 2017, doi: 10.1016/j.oceaneng.2017.06.020

[20] R. Ghimire, C. Zhang, and K. R. Pattipati, “A Rough Set-Theory-Based

Fault-Diagnosis Method for an Electric Power-Steering System,”

IEEEASME Trans. Mechatron., 2018, doi: 10.1109/TMECH.2018.2863119

[21] P. Lezanski and M. Pilacinska, “The dominance-based rough set approach

to cylindrical plunge grinding process diagnosis,” J. Intell. Manuf., Vol. 29,

No. 5, pp. 989-1004, Jun. 2018, doi: 10.1007/s10845-016-1230-1


– 249 –

[22] A. Soualhi, G. Clerc, and H. Razik, “Detection and Diagnosis of Faults in

Induction Motor Using an Improved Artificial Ant Clustering Technique,”

IEEE Trans. Ind. Electron., Vol. 60, No. 9, pp. 4053-4062, Sep. 2013, doi:

10.1109/TIE.2012.2230598

[23] B. M. Ebrahimi, M. Javan Roshtkhari, J. Faiz, and S. V. Khatami,

“Advanced Eccentricity Fault Recognition in Permanent Magnet

Synchronous Motors Using Stator Current Signature Analysis,” IEEE

Trans. Ind. Electron., Vol. 61, No. 4, pp. 2041-2052, Apr. 2014, doi:

10.1109/TIE.2013.2263777

[24] J. K. Scott, G. R. Marseglia, L. Magni, R. D. Braatz, and D. M. Raimondo,

“A hybrid stochastic-deterministic input design method for active fault

diagnosis,” in 52nd IEEE Conference on Decision and Control, Dec. 2013,

pp. 5656-5661, doi: 10.1109/CDC.2013.6760780

[25] Y. Song and B. Wang, “Survey on Reliability of Power Electronic

Systems,” IEEE Trans. Power Electron., Vol. 28, No. 1, pp. 591-604, Jan.

2013, doi: 10.1109/TPEL.2012.2192503

[26] S. Yang, A. Bryant, P. Mawby, D. Xiang, L. Ran, and P. Tavner, “An

Industry-Based Survey of Reliability in Power Electronic Converters,”

IEEE Trans. Ind. Appl., Vol. 47, No. 3, pp. 1441-1451, May 2011, doi:

10.1109/TIA.2011.2124436.

[27] G. Györök, “The FPAA realization of analog robust electronic circuit,” in

2009 IEEE International Conference on Computational Cybernetics

(ICCC), Jan. 2009, pp. 179-183, doi: 10.1109/ICCCYB.2009.5393941

[28] P. Holcsik, J. Pálfi, Z. Čonka, and M. Avornicului, “A Theoretical

Approach to The Implementation of Low-Voltage Smart Switch Boards,”

Acta Polytech. Hung., Vol. 16, No. 4, Jul. 2019, doi:

10.12700/APH.16.4.2019.4.7

[29] G. T. Orosz, A. Sulyok, G. Gergely, S. Gurbán, and M. Menyhard,

“Calculation of the Surface Excitation Parameter for Si and Ge from

Measured Electron Backscattered Spectra by Means of a Monte-Carlo

Simulation,” Microsc. Microanal., Vol. 9, No. 4, pp. 343-348, Aug. 2003,

doi: 10.1017/S1431927603030241

[30] G. T. Orosz et al., “Experimental determination of electron inelastic

scattering cross-sections in Si, Ge and III-V semiconductors,” Vacuum,

Vol. 71, No. 1, pp. 147-152, May 2003, doi: 10.1016/S0042-

207X(02)00729-7

[31] G. Gergely et al., “Surface excitation correction of the inelastic mean free

path in selected conducting polymers,” Appl. Surf. Sci., Vol. 252, No. 14,

pp. 4982-4989, May 2006, doi: 10.1016/j.apsusc.2005.07.017

a highly reliable, modular, redundant and self-monitoring psu...

Documents